BLIND ROBUST AUDIO WATERMARKING BASED ON REMAINING NUMBERS IN DISCRETE COSINE TRANSFORM

The rapid growth and increasing of information technology and computer network have caused that data transition in digital form is considered more and more. The main disturbance of owners and producers of digital products is to protect distribution and unauthorized copying. An effective solution for this problem is digital watermarking techniques. The purpose of audio digital watermarking is to insert a series of hidden information into audio file so that it is not heard and robust against signal processing attacks. This paper proposes a new audio watermarking method that is able to embed watermark sequence into host signal based on remaining numbers. The extraction process is carried out blindly without any knowledge about original audio signal. For comprehensive evaluation of the proposed method, several experiments including subjective and objective tests (MOS, SNR and SNRseg measures), robustness and performance (BER, NC, TP and FP measures) against Stirmark attacks as well as capacity have been performed. Experimental results show that the proposed method is able to improve trade-off among three measures transparency, robustness, and capacity.